164 THE ROYAL SOCIETY OF CANADA 



The compound curve is shown in dotted lines in the figure and 

 it is quite evident that it does not represent the observations much 

 if any more satisfactorily than the simple curve, for, while the agree- 

 ment is better between phases 80 and 130, it is poorer at other parts. 

 The average plate residual is only reduced about 5% by the intro- 

 duction of the secondary. 



One possible explanation of the deviations is the presence of the 

 spectrum of the companion to the principal star and the displacement 

 of the measured velocities towards the 7 line by the blend effect of the 

 two spectra. 



It is difficult to see how such a blend effect can cause deviations 

 of the peculiar character shown here, as the curve goes through a 

 complete cycle below the 7 line and exhibits no evidence of blending 

 above this line. Yet Harper's orbit of Aquilae 1 , a binary whose 

 elements are quite similar to those of # 2 Tauri, shows similar deviations 

 below the 7 line though not so strongly marked as here. 



In the case of # Aquilae, it was later shown that this was due 

 to the presence of a second spectrum with the resultant blend effect. 



The inference is that the second spectrum is present in # 2 Tauri ; 

 but as yet no reasonable evidence to that effect has been secured. 

 On two of the early plates obtained at Yerkes, the second spectrum 

 was measured and on four obtained here some apparent doubling was 

 observed. The results at Yerkes and the attempted measures here, 

 given in the last column of the table of measures, however all bring 

 the secondary spectrum in impossible positions. For example, the 

 secondary velocity in one plate and the primary velocity in the other 

 plate at Yerkes fall within two or three kilometre of the 7 line, while 

 the velocities of primary and secondary in every suspected case here are 

 in equally impossible relative positions. Furthermore, later trials on 

 these suspected plates found me unable to repeat my measures and I 

 strongly doubt the reality of the doubling. 



As previously stated, plates have been especially secured here near 

 the maximum positive velocity, when the doubling should be more 

 pronounced, with three different dispersions; but in no case can doubled 

 lines be definitely seen, and while there is possibly a second spectrum 

 present the lines are so broad diffuse and lacking in contrast that I 

 doubt whether it is possible to definitely settle the question. 



Other reasons may be cited for suspecting abnormal conditions 

 in this star. The large residuals from this orbit of some of the plates 

 obtained at the Lick and Yerkes Observatories, the average residual 

 being 8 • 4 as compared with 4 • 1 km at Ottawa, are much greater than 

 can be explained by the poor character of the spectrum or by the 



journal R.A.S.C. Ill p. 87, Man-April, 1909. 



